Octylamine‐Supporting Interlayer Expanded Molybdenum Diselenide as a High‐Power Cathode for Rechargeable Mg Batteries

Rechargeable Mg batteries (RMBs) are a promising large‐scale energy‐storage technology with low cost and high safety, but the performance is limited by the inferior kinetics of Mg‐intercalation cathodes. In the present study, an octylamine‐supporting interlayer expanded molybdenum diselenide (e‐MoSe2) is synthesized and used as cathode for RMBs, in comparison with ordinary crystalline MoSe2. The octylamine molecules introduced show a strong interaction with the MoSe2 layers and increase the layer spacing significantly from 6.46 to 11.5 Å. e‐MoSe2 shows a high Mg‐storage capacity of 238 mAh g−1 at 50 mA g−1 and a superior rate performance of 39 mAh g−1 at 10 A g−1, far advantageous over crystalline MoSe2. e‐MoSe2 also shows a considerably high structure stability during repeated magnesiation/demagnesiation, providing an outstanding cycling stability for 1000 cycles. Further electrochemical tests demonstrate the high Mg2+ diffusion coefficients in e‐MoSe2. Theoretical computation indicates the interlayer expansion changes the Mg2+ diffusion paths from “hollow site → hollow site” to “hollow site → Se atom site → hollow site”, largely decreasing the energy barrier and improving the Mg2+ diffusion kinetics. The present work highlights an efficient strategy for the improvement of Mg‐storage performance for RMB cathodes.


Introduction
As the most representative electrochemical energy storage technology, Li-ion batteries (LIBs) have dominated the markets of portable electronic devices and electric vehicles with the high energy density and long life span.However, their shortcomings in cost and safety hinder the application for grid energy storage.Therefore, many types of novel battery systems are currently being explored for potential candidates. [1]epresentatively, rechargeable Mg batteries (RMBs) exhibit unique features because of the special Mg anode.4] More importantly, Mg anode shows a low tendency to form dendrites in the electrochemical deposition, thus ensuring high safety for applications. [5,6]Despite the merits, the major hindrance of RMB development lies in the exploration of suitable cathode materials.Although with a similar radius (0.86 A) as Li + cation (0.90 A), the Mg 2+ cation have two units of positive charge.The high charge density makes the Mg 2+ intercalation into the narrow inorganic crystal lattice intrinsically difficult, leading to low capacities and sluggish Mg 2+ diffusion kinetics. [7,8]ayered compounds with 2D tunnels are promising host materials for guest cations such as Li + , Na + and K + , which have also been attempted as RMB cathodes, such like V 2 O 5 , [9][10][11] VOPO 4 , [12] VS 2 , [13] MoS 2 , [14] WSe 2 [15] and NbSe 2 . [16]However, due to the strong interaction between the host and bivalent Mg 2+ cation as well as unfavorable Mg 2+ diffusion paths, the Mg-storage performances of such materials are inferior compared with the storage of monovalent cations, in terms of low capacity, poor rate capability and slow Mg 2+ diffusion kinetics. [17]To address this issue, an effective strategy is to expand the interlayer spacing, providing favorable Mg 2+ diffusion routes, weakening the interaction and thus facilitating the Mg 2+ transport. [18]Usually, the substance introduced into the layer could be either metallic cations (Na + , K + , [19] Mg 2+ [20] and Mn 2+ [21] ) or organic molecules (phenylamine, [22] poly (ethylene oxide) [23] and 2-ethylhexylamine [24] ).For example, Mai et al. reported that pre-intercalation of phenylamine into VOPO 4 layers (PA-VOPO 4 ) could largely increase the layer spacing from 7.4 A to 14.2 A and effectively improve the electrochemical performance.PA-VOPO 4 exhibits an outstanding rate capability (109 mAh g À1 at 2.0 A g À1 ) and a superior cycling stability (500 cycles). [22]Jin et al. synthesized interlayer expanded VS 2 nanoflowers by pre-intercalation of 2-ethylhexylamine, which shows an excellent rate performance of 103 mAh g À1 at 2.0 A g À1 . [24]s a high-performance interlayer-expanded 2D Mg-storage host, the material should have a stable structure and weak interaction with the Rechargeable Mg batteries (RMBs) are a promising large-scale energy-storage technology with low cost and high safety, but the performance is limited by the inferior kinetics of Mg-intercalation cathodes.In the present study, an octylamine-supporting interlayer expanded molybdenum diselenide (e-MoSe 2 ) is synthesized and used as cathode for RMBs, in comparison with ordinary crystalline MoSe 2 .The octylamine molecules introduced show a strong interaction with the MoSe 2 layers and increase the layer spacing significantly from 6.46 to 11.5 A. e-MoSe 2 shows a high Mg-storage capacity of 238 mAh g À1 at 50 mA g À1 and a superior rate performance of 39 mAh g À1 at 10 A g À1 , far advantageous over crystalline MoSe 2 .e-MoSe 2 also shows a considerably high structure stability during repeated magnesiation/ demagnesiation, providing an outstanding cycling stability for 1000 cycles.Further electrochemical tests demonstrate the high Mg 2+ diffusion coefficients in e-MoSe 2 .Theoretical computation indicates the interlayer expansion changes the Mg 2+ diffusion paths from "hollow site ?hollow site" to "hollow site ?Se atom site ?hollow site", largely decreasing the energy barrier and improving the Mg 2+ diffusion kinetics.The present work highlights an efficient strategy for the improvement of Mg-storage performance for RMB cathodes.
bivalent Mg 2+ cation. [25]Compared with O 2À and S 2À , Se 2À is advantageous because of its larger size and weaker interaction with the Mg 2+ cation. [26]While for the cations, Mo, Ti and V are better selections. [27,28]Because of the similar energy levels of the Mo 4d (Ti 3d or V 3d) and Se 4p orbitals, Mo forms covalent-like bonds with Se anion with orbital overlap. [27]Therefore, the layer structure stability is largely enhanced and the negative charge is effectively delocalized to the whole layer, hence its interaction with Mg 2+ is significantly weakened.In the present study, an octylamine-supporting interlayer expanded MoSe 2 (e-MoSe 2 ) is synthesized and used as cathode for RMBs.Unlike cations which may be extracted during cycling, the neutral octylamine molecules would not migrate in electric field.More significantly, the octylamine molecule shows a strong interaction with the MoSe 2 layer, which enhances the structure stability and improves the cyclability.e-MoSe 2 exhibits a high Mg-storage capacity of 238 mAh g À1 and an outstanding cycling stability for 1000 cycles.The bulky octylamine molecule expands the MoSe 2 layer spacing significantly from 6.46 to 11.5 A, leading to a decreased diffusion energy barrier and fascinating rate performance of 39 mAh g À1 at 10 A g À1 .The present study develops a high-power cathode material for RMBs, and highlights an efficient strategy for the improvement of Mg-storage performance for RMBs cathodes.

Results and Discussion
Figure 1 describes the preparation of e-MoSe 2 and the operation principle of the RMBs.In the solvothermal process, octylamine acts as not only a cosolvent to dissolve MoO 3 and SeO 2 via complexes, but also an agent reducing SeO 2 to Se 2À for the subsequent construction of MoSe 2 .Moreover, the octylamine molecules are distributed among the MoSe 2 layers, functioning as support for the ultra-expanded layer spacing.Figure S1, Supporting Information, shows the XRD results of the products with different solvothermal times.After a solvothermal reaction of 1 h, all SeO 2 was reduced to Se by octylamine, which would aggregate and grow directionally to form the unique morphologic substrates.After 3 h of reaction, part of Se transformed to e-MoSe 2 (Figure S1, Supporting Information).After 6 h, the second reaction step was completed, notified as the disappearance of the Se peaks (Figure S1, Supporting Information).Such a two-step reduction process slows down the formation of MoSe 2 and provides sufficient time for the formation of interlayer expanded e-MoSe 2 and the growth of special nanotube morphology (Figure 2).
The morphology of e-MoSe 2 is shown in Figure 2. The SEM images describe that e-MoSe 2 is hollow nanotubes with length of 0.5-2.0lm (Figure 2a).It is also observed that the nanotubes have a coarse surface and are composed of interconnected nanosheets (Figure 2b), which is also demonstrated by the TEM observation (Figure 2c,d, diameter of about 250 nm).Such morphology delivers large electrode/electrolyte interphase and short-range diffusion paths, which would improve the interphase charge transfer and solid-state Mg 2+ transport.HRTEM reveals that e-MoSe 2 exhibits plenty of uneven fringes, indicating the low crystallinity (Figure 2e).The spacing of the fringes in different regions fluctuates in the range of 6.8-11.5A (Figure S2, Supporting Information).Figure 2f shows the fringe of 11.5 A, which is much larger than the original MoSe 2 (6.46 A) and would greatly reduce the diffusion hindrance of Mg 2+ .As shown in Figure 2g, two wide diffraction rings (r1 and r2) appeared in the SAED image, corresponding to the two broad XRD peaks (Figure 3a) and indicating the low crystallinity.The EDS element mapping of C and N demonstrates the uniform distribution of octylamine molecules in the e-MoSe 2 nanotubes (Figure 2h).SEM was also used to observe the morphology of a-MoSe 2 and b-MoSe 2 (Figure S3).The morphology of a-MoSe 2 does not change much after calcination of e-MoSe 2 in Ar atmosphere, still maintaining the hollow nanotube structure, which makes the two samples good comparison counterparts.As for commercial b-MoSe 2 , it appears as broken nanosheets and nanoparticles after 1 h of ball milling.
The atomic ratio of Mo to Se in e-MoSe 2 was measured via ICP-MS, and the result is 0.93: 2.00, generally in agreement with the formula of MoSe 2 .Figure 3a shows the XRD patterns of e-MoSe 2 , a-MoSe 2 and b-MoSe 2 .The sharp peaks of b-MoSe 2 at 13.7°, 27.6°, 31.4°and34.4°c orrespond to the (002), (004), (100) and (102) planes, respectively, in good accordance with MoSe 2 (JCPDS No.29-0914).Two broad peaks of e-MoSe 2 appear at 33°and 55°, indicating its low crystallinity and consistent with the SAED results (see r1 and r2 in Figure 2g).No characteristic peaks corresponding to (002) crystal plane was observed for e-MoSe 2 in the low angle region (see Figure S4, Supporting Information, for the small angle XRD).According to HRTEM, this could be attributed to the fluctuation of the layer spacing (Figure S2, Supporting Information).Amorphous or low-crystalline materials could exhibit improved Mg-storage performance, which is another factor for e-MoSe 2 in addition to interlayer expansion. [29]he peaks of a-MoSe 2 are much sharper than those of e-MoSe 2 , indicating the greatly improved crystallinity upon calcination.The characteristic (002) peak is present at 13.2°, and the layer spacing decreases (6.7 A) almost to the pristine MoSe 2 (6.46 A) due to the removal of octylamine molecules during calcination.The A 1g peak at 245 cm À1 and E 2g peak at 280 cm À1 in the Raman spectra of the three samples were attributed to the Mo-Se bond (Figure S5, Supporting Information). [30]  which are identical as the peaks of octylamine (Figure 3b). [31]This demonstrates the existence of octylamine in e-MoSe 2 .The absorption band at 1403 cm À1 is ascribed to a symmetric bending of the N-H bond caused by the -NH 2 group in the octylamine molecule. [32]he broad peak at 990 cm À1 is attributed to the -OH group on the surface of the amorphous carbon, which might be generated during the solvothermal synthesis.In contrast, the IR spectrum of a-MoSe 2 is similar with that of b-MoSe 2 and does not contain the octylamine peaks, indicating removal of the octylamine molecules during calcination.The TG curves of e-MoSe 2 and octylamine are compared in Figure 3c.The weight loss of e-MoSe 2 occurs at 300 °C (~20 wt.%) and 450 °C (~38 wt.%).In sharp contrast, the weight of pure octylamine is completely lost at 150 °C.This evidence demonstrates the strong interaction between octylamine molecules and MoSe 2 layers in e-MoSe 2 , and such a strong interaction is important for the structure stability of e-MoSe 2 against long-term cycling as RMB cathodes.
Figure 3d-f shows the XPS spectra of the three samples.In the Mo 3d spectra (Figure 3d), both of a-MoSe 2 and b-MoSe 2 exhibit a pair of double peaks at 231.8 and 228.6 eV, respectively corresponding to the Mo 3d 3/2 and Mo 3d 5/2 orbitals of Mo 4+ in MoSe 2 . [33]Another two pairs of peaks are present for e-MoSe 2 at 234.3 and 231.1 eV, 235.9 and 232.7 eV, which respectively belong to Mo 5+ and Mo 6+ . [34]They are ascribed to the surface oxidation, which is quite common for synthesized MoSe 2 .Figure 3e shows the Se 3d spectra.Similar peaks corresponding to Se 3d 3/2 and Se 3d 5/2 orbitals appear at 55.1 and 54.3 eV in all the three samples, suggesting that Se exists in the form of Se 2À . [35]In the N 1s spectra (Figure 3f), besides the Mo 3p peaks, e-MoSe 2 also shows a large signal at 398.4 eV, which is attributed to -NH 2 of octylamine.A small peak at 401.3 eV is speculated to be trace amount of NH 4 + generated during the solvothermal synthesis. [36]The specific surface area was measured to be 9.38 m 2 g À1 (Figure S6, Supporting Information).Then e-MoSe 2 , a-MoSe 2 and b-MoSe 2 were investigated as cathodes for RMBs.Mg(TFSI) 2 -MgCl 2 electrolyte was used because it is highly compatible with the metallic Mg anode (Figure S7, Supporting Information). [37]First, CV curves were measured for e-MoSe 2 (Figure S8, Supporting Information and Figure 4c).The first cycle is different from the following ones because of the modulation of the Mg anode surface and the cathode materials.Reproducible CV curves were obtained after initial cycles (Figure 4c), indicating the highly reversible magnesiation/demagnesiation process.The reduction peak at 1.15 V and oxidation peaks at 1.55 and 2.05 V suggest the one-step magnesiation and two-step demagnesiation processes.][40][41][42][43][44] The fast Mg 2+ diffusion kinetics of e-MoSe 2 was also verified by CV tests, in which the CV profiles almost remain the same shape at different scan rates (Figure S12a, Supporting Information).The Mg 2+ diffusion coefficients are determined to be 3.8-5.29 10 À10 cm 2 s À1 (see Table S1, Supporting Information, for details), which are comparable to copper sulfides and selenides with unique displacement mechanism and two-magnitude higher than those of a-MoSe 2 (1.4-2.1 9 10 À12 cm 2 s À1 ).The long-term cycling at 1.0 A g À1 also demonstrates the outstanding cyclability of e-MoSe 2 over 1000 cycles (Figure 4g, see Figure S13a, Supporting Information, for the charge/discharge curves at different cycles).The cycled e-MoSe 2 still maintained the nanotube morphology (Figure 4h and Figure S13b, Supporting Information).The hierarchical nanotube morphology could accommodate the structure change during cycling.More importantly, the strong interaction between octylamine molecules and MoSe 2 layers confirms the structure stability against long-time magnesiation/demagnesiation, resulting in a stable cycling.The EIS results are given in Figure S14, Supporting Information.The impedance of e-MoSe 2 is lower than those of a-MoSe 2 and b-MoSe 2 at the 5th cycle (Figure S14a, Supporting Information), and did not increase obviously after 1000 cycles (Figure S14b, Supporting Information).The Mg-storage mechanism was investigated via ex-situ XRD, XPS and TEM.As seen in the XRD patterns (Figure 5a), no sharp peaks is observed except for the peaks from the PTFE binder (18.1°) and carbon cloth current collector (25.8°, 43.1°and 54.3°), indicating the low crystallinity of e-MoSe 2 all through the cycling.In the Mo 3d XPS spectra (Figure 5b), The peaks corresponding to Mo 4+ weakened at the discharge states and recovered to the original level at the charge states (both of first and 15th cycles), indicating the redox of Mo between Mo 4+ and lower valence state.However, lowvalence-state Mo was not found in the discharge state, but the peaks of MoO 3 are significantly enhanced at 236.4 and 233.2 eV instead.That is because of the instability of low-valence-state Mo, which is easily to be oxidized in the air. [45]The valence state of Se does not change obviously and remains Se 2À during the cycling (Figure 5c).While in the Mg 1s spectra (Figure 5d), the peak increases/decreases at the discharge/charge states, corresponding to the intercalation/deintercalation of Mg 2+ . [46]Comparatively, the peak intensity at the 15th discharge state is slightly higher than that of the first discharge state, which suggests this irreversible magnesiation could be the reason for the capacity fading during the first 15 cycles (Figure 4b).
The e-MoSe 2 powder was scraped off the cycled electrodes (charge and discharge states) and sent for ex-situ TEM tests.As shown in Figure 6a,b, the hollow nanotube structure and uneven lattice stripes are still clearly visible, along with the small particles scattered around (conductive carbon KB).As shown in Figure 6c,d, the maximum layer spacing was slightly reduced to 10.2 A at the charge state after 15 cycles.However, unlike calcination at high temperature, the layer spacing did not return to the original level.The electrochemical performance verifies the layer spacing of 10.2 A large enough to accommodate Mg 2+ for fast intercalation/deintercalation. Figure 6e-g are the HRTEM images and fringes of the discharged powder for the lattice spacing measurements in different regions.It is clearly seen that the Mg-intercalation into the layers increases the layer spacing to 10.6 A. Figure 6h schematically describes the structural changes of e-MoSe 2 after cycling according to the above results.
Theoretical computation was further conducted to explain the superior kinetics of e-MoSe 2 .The Mg 2+ diffusion paths and energy barriers of b-MoSe 2 (crystalline) and e-MoSe 2 (interlayer expanded) are described in Figure 7.As shown in Figure 7a,b, Mg 2+ can only locate at the hollow sites between the layers in the ordinary crystalline MoSe 2 , which makes Mg 2+ could only transport between the adjacent hollow sites (Figure S15, Supporting Information).The energy barrier is 0.432 eV for such a diffusion path.While the interlayer spacing increases from 6.46 to 10.2 A (e-MoSe 2 ), Mg 2+ can also be adsorbed at the Se atom sites between the two layers.This changes the Mg 2+ diffusion paths from "hollow site ?hollow site" (Figure 7b) to "hollow site ?Se atom site ?hollow site" (Figure 7d and Figure S16, Supporting Information).A possible evidence for this is there are two charge plateaus for e-MoSe 2 , which might be owing to the dissociation of Mg 2+ from different sites, while there is only one plateau for a-MoSe 2 .Correspondingly, the Mg 2+ diffusion energy barrier decreases significantly from 0.432 eV (Figure 7a) to 0.167 eV (Figure 7c).The increase of interlayer adsorption sites and the decrease of diffusion energy barrier are the origin of the excellent kinetic performance of e-MoSe 2 . [47]For the efficient storage of bivalent Mg 2+ , the interlayer extension and disorder arrangement of the layers are advantageous over ordered crystalline layered compounds.Such extended layer spacing and disorder arrangement construct a considerably open structure for Mg 2+ intercalation and diffusion as well as plenty of Mg 2+ adsorption sites, effectively reducing the energy barrier and accelerating the kinetics.It is also highlighted that this could be a promising strategy for the design of high-performance RMB cathode materials.

Conclusions
In summary, an octylamine-supporting interlayer expanded MoSe 2 (e-MoSe 2 ) is synthesized and investigated as RMBs cathode material in comparison with annealed e-MoSe 2 (a-MoSe 2 ) and commercial MoSe 2 (b-MoSe 2 ).The octylamine molecules introduced increase the layer spacing significantly from 6.46 to 11.5 A and largely decrease the crystallinity.The annealed e-MoSe 2 (a-MoSe 2 ) shows a layer spacing similar with commercial MoSe 2 (b-MoSe 2 ) and a much higher crystallinity than e-MoSe 2 upon removal of octylamine.e-MoSe 2 exhibits a high Mg-storage capacity of 238 mAh g À1 at 50 mA g À1 and a superior rate capability of 39 mAh g À1 at 10 A g À1 , far advantageous over a-MoSe 2 and b-MoSe 2 .The octylamine molecules introduced into e-MoSe 2 show a strong interaction with the MoSe 2 layers, enhancing the structure stability and leading to an outstanding cycling stability for 1000 cycles.Further electrochemical tests demonstrate the Mg 2+ diffusion coefficients of e-MoSe 2 two-magnitude higher than those of a-MoSe 2 .Theoretical computation indicates the interlayer expansion changes the Mg 2+ diffusion paths from "hollow site ?hollow site" to "hollow site ?Se atom site ?hollow site", significantly decreasing the energy barrier and improving the Mg 2+ diffusion kinetics.The principles revealed herein highlight an effective strategy for the improvement of Mg-storage performance and the realization of high-power RMBs cathodes.

Electrochemical measurements:
The electrode material slurry contained the cathode material, conducting carbon (KB) and binder (PTFE) in a mass ratio of 6:3:1.
Isopropyl was used as the solvent for the slurry.The slurry was evenly coated on the current collector (carbon cloth) and dried at 60 °C for 1 h in a vacuum oven.Finally, the dried electrodes were cut into small discs (diameter of 12 mm).The active material loading was about 1.5 mg cm À2 .The test cell is fabricated in the Ar glove box with the prepared cathodes, polished metallic Mg anodes, Mg electrolyte and Whatman separator (Whatman, GF/A).CR2032 coin cell was used with a Mo plate inserted to prevent electrochemical corrosions.The Mg electrolyte was an anhydrous DME (10 mL, dehydrated with activated 3 A molecular sieves) solution containing Mg(TFSI) 2 (2.5 mmol, Sigma aldrich, 97%) and MgCl 2 (5.0 mmol, Alfa asear, 99%) dissolved therein.The LAND testers were used to measure the galvanostatic charge/discharge and rate performance of the cell.Cyclic voltammetry (CV) was tested with Chenhua CHI 660e electrochemical workstation.The electrochemical impedance spectroscope (EIS) was measured by CS310 workstation (Coster).The voltage amplitude was 10 mV, and the frequency was from 10 5 to 0.1 Hz.All the electrochemical tests were conducted at 25 °C.For ex-situ XRD and XPS characterizations of electrodes, cells at different charge/discharge states were disassembled in the glove box, and the electrodes were washed with anhydrous DME, and then sent for testing.As for the ex-situ TEM characterization, the electrode materials were scraped out of the cycled electrodes and sent for the TEM observation.
Computational details: Calculation was conducted with density functional theory (DFT) and projector augmented wave method, implemented in Vienna ab-initio simulation package (VASP). [48]Generalized gradient approximation (GGA) was used to describe the exchange correlation energy.The cut-off energy was set as 450 eV for the plane wave basis.The energy criterion was set to 10 À7 eV for the iterative solution of the Kohn-Sham equation.All structures were relaxed.For optimization of MoSe 2 , a (3 9 2 9 1) supercell was used.The Brilouinzone integration was conducted with a (2 9 3 9 1) Monkhorst Pack grid.The Mg 2+ diffusion of the two models of MoSe 2 was analyzed with climbing image nudged elastic band (CI-NEB) method. [49]Seven images were used for each Mg 2+ diffusion path.The difference between the highest and lowest energies was defined as the diffusion barrier.
Figure 3b compares the FTIR spectra of the three samples and octylamine.The spectrum of e-MoSe 2 is obviously different from those of a-MoSe 2 and b-MoSe 2 .The absorption peaks of e-MoSe 2 at 2924 and 2852 cm À1 are attributed to the vibration of -CH 3 or -CH 2 groups,

Figure 1 .
Figure 1.The preparation of e-MoSe 2 as the cathode material for RMBs.

Figure 3 .
Figure 3. a) XRD patterns, b) FTIR spectra and XPS spectra of d) Mo 3d, e) Se 3d and f) N 1 s for e-MoSe 2 , a-MoSe 2 and b-MoSe 2 .c) TG curves of e-MoSe 2 and octylamine in air.

Figure 5 .
Figure 5. Ex-situ a) XRD patterns, XPS spectra of b) Mo 3d, c) Se 3d and d) Mg 1s of e-MoSe 2 electrodes at different states.

Figure 6 .
Figure 6.a-c) TEM images and d) lattice fringe of e-MoSe 2 at the charge state after 15 cycles.e, f) HRTEM images and g) lattice fringe of e-MoSe 2 at the discharge state after 15 cycles.h) Scheme for structural changes of e-MoSe 2 .
The mixture was transferred to a 100 mL solvothermal reactor and kept at 200 °C for 8 h.The black precipitate in the solution was collected and washed repeatedly with distilled water and ethanol.Finally, the e-MoSe 2 product was vacuum-dried at 60 °C for 12 h.The sample of a-MoSe 2 was obtained via calcination of e-MoSe 2 at 400 °C for 3 h under flowing Ar.Commercial MoSe 2 (b-Mose 2 , Macklin, 99%) sample was also studied for comparison.b-MoSe 2 was subjected to planetary ball milling for 1 h before use.
Characterization:The structures of the samples were analyzed by a D8 ADVANCE X-ray diffractometer (XRD) using Cu Ka radiation at a scan rate of 5°min À1 .Thermo's ESCALAB 250Xi X-ray photoelectron spectrometer (XPS) was performed to analyze the elements and valence states, and the binding